Concealed Questions Are Structured

Questions in natural language typically consist of a proposition articulated into a foreground part, which evokes a set of alternatives, and a background part, which provides a function for distinguishing among those alternatives. For instance, the question Who left? asks which of the people that who quantifies over (the foreground) have the property of leaving (the background). Concealed questions are DPs that can be interpreted as if they were interrogatives. For instance, I found out Bill’s age means roughly the same thing as I found out what Bill’s age is, so the DP Bill’s age is functioning here as a concealed question. It has long been observed by Löbner and others that DPs containing relational nouns such as age make excellent concealed questions, but simple DPs headed by sortal nouns such as brick do not: it is difficult or impossible to interpret ??I found out the brick as meaning the same thing as I found out which object the brick is. I propose that, just like ordinary clausal questions, concealedquestion DPs must be articulated into a foreground and a background. For a DP headed by a relational noun, such as Bill’s age, the possible referents serve as the set of alternatives (that is, the set of ages), and the background property corresponds to standing in the relation denoted by the head noun to the possessor (Bill). In contrast, simple sortal DPs provide only a set of alternatives (e.g., the set of bricks), without providing any non-trivial property for distinguishing one brick from another. It is this lack of foreground/background structure that prevents DPs headed by sortal nominals from serving as concealed questions. If so, then the denotations of definite descriptions must have internal structure in the way that has long been argued for propositions. 1. Concealed questions and the sortal/relational 4